Could anyone give me some advice, please?

I'm pretty sure this caneffectively be done, but I'm wondering if anyone has
any advice on whether its a good way to do it, or what problems to look out
for - that kind of thing?

Basically, I am replacing the current system of VBA code driving the Word
2002 object model to build documents from autotext and populate data into
bookmarked areas, etc. with a .NET/Word 2007 solution.
My idea for the new system is to record components in building blocks like
the old autotext, but use content controls to identify the components and the
data fields within them then use a mix of .NET code automation and XSLTs
working on the WordML of the building blocks to build the document.

Simplified, I suppose I would, in Word 2007, 'design' the document by
recording all the BB components and carefully tag them with CCs then store
them in a central location (probably put the template holding them in
Sharepoint), then have the 'rendering' code retrieve the template and attach
it to the new (empty) document and, according to configuration data dictating
the document make-up and data sources, etc., pull them all in to make the
document.

It would be nice if perhaps components could be built using just XSLTs to
combine data XML and BB WordML and then insert the output XML into the
document directly without automating the Word object model, but I'm thinking
in reality a lot of XPath and Word automation will be needed.

I'd love to hear from anyone who has tried this!

Thanks very much in advance for any advice.

--
Big Dave UK x

Hi Dave,

> I'd love to hear from anyone who has tried this!
>
You're more likely to get a discussion going in a
developer-level group than this one, targeted at end-users
:-)

Have you considered building the document completely using
OpenXML?

I'm thinking more and more that a VSTO solution might be a
good bet. But you'd have to wait for Orcas to do that with
2007.

Cindy Meister